JLH 10 Watt class A amplifier

Next I increase the current to 1 amp. After which something must have blown. I didn't hear or smell anything but I could't read any voltage on the output terminals any longer.

Nothing bad happened, no smell or anything. With a dual-rail amplifier like this version the output should be close to zero volts.

So I spent a lot of time checking voltages according to the chart you gave me. Unfortunately everything seems to be in order. Nothing obvious that jumps out at me.

The only thing said so far that is troublesome is that the output devices didn't get warm and that without voltage across R10 it suggests the amp hasn't been adjusted to set the output bias. The output bias current is constrained by the upper output device (Q2) (the negative feedback of the amplifier will ensure that the lower output device has the same d.c. current as the upper output device). The dc current through Q2 is constrained by the current flowing into it's base - which you adjust with VR2. A fixed value resistor in series with VR2 will make it much easier to measure the current through VR2.

By the way, the output devices may not get instantly get warm, takes some time depending on size of heatsink. On my amplifier it takes a few minutes before they get warm and a substantial time before the whole thing is at running temperature.
 
Last edited:
Voltages

Ok, I swapped out VR1 and VR2. Powered up and took the following voltage readings using 22v on the rails and 0.01 amps:

An externally hosted image should be here but it was not working when we last tested it.


There is something going on at Q4. From the base of Q4 through R8 to Vout I should be read millivolts, instead I'm getting 5v at Q4 base and 22v on R8. Also when I take the reading at the base of Q4 the relays in my power supply go crazy meaning they are flipping back and forth between reading voltage and current.

I'm also getting 2.2v at the base of Q2 instead of ~539mv.

I also measured across B-E on all transistors and got the following readings:

Q1 - 21.4v
Q2 - 97mv
Q3 - 0v
Q4 - 0v
Q5 - 0v
Q6 - .33v
Q7 - 0v
Q8 - .33v

I retraced all my connections and as far as I can tell I've hooked everything up correctly...

Thanks for your help -
Bruce
 
Edit: do number 3/ first.

Let's look at where the current is flowing - or where it is not.

Something is wrong with the two-transistor current source feeding the phase splitter - i.e. I'm looking at Q7 and Q8.

If you have 22V across R12 then you have no voltage drop across Q7 - this is wrong, should be some voltage drop there.

1/ Check that the bottle-end of R12 is indeed connected to ground - if not this could explain many things.
2/ Check that R12 is not open circuit, best if you can check it is in fact 4k7 and not some much higher value (the colour coding on the resistor maybe wrong - it's not been unknown to happen).

You also have 22V across R6, which would imply 100mA of current flow and I don't see where that is coming from.

3/ Review all your ground connections - where does the ground trace back to the power supply - what does your power supply look like. Could be that your ground is open-circuit somewhere and so no current can flow
 
Last edited:
I don't believe your measurements abpea. Your B-E measurements do not tally with what you indicate on the circuit. Go do the measurements again and make sure that your meter negative is connected to the zero volt (centre tap) of the power supply.

Gareth, before analyzing the problem let him remeasure, there is more than meets the eye going on here.
 
There is no way this circuit is going to be dc stable without a capacitor in series with R6. The voltage on the emitter of Q4 is amplified ~x13.

Those pots VR1 and VR2 need some modification to make the circuit less likely to self destruct when bumped and what is VR3 meant to achieve?

That may just solve the problem. Also VR2 is more likely to be about 47 ohms for 1 amp to be sourced by Q2.

Edit: The current through Q6 should be about 730 uA and therefor VR1=877 Ohm
 
Last edited:
Administrator
Joined 2007
Paid Member
I don't believe your measurements abpea. Your B-E measurements do not tally with what you indicate on the circuit.

Yes, something isn't right here. If you measure 22 volts on Q4 emitter and 5 volts on the base, then the voltage across B and E has to be in the region of 17 volts (which would indicate it being open circuit.

There is no way this circuit is going to be dc stable without a capacitor in series with R6. The voltage on the emitter of Q4 is amplified ~x13.

My original simulation file for this amp had such a cap in place... I can't remember how I came to have created the file tbh, but it must be related to an earlier post in this thread. It does work without though although the DC stability will be, erm, interesting.

Those pots VR1 and VR2 need some modification to make the circuit less likely to self destruct when bumped and what is VR3 meant to achieve?

Maybe some attempt at temperature compensation... hard to say really because at face value it does little.

Here is the file if anyone wants a play.
 

Attachments

  • Capture1.PNG
    Capture1.PNG
    183 KB · Views: 414
  • JLH 2005 Class A Amp.asc
    4.5 KB · Views: 65
I built the amp on a breadboard, point-to-point. I have the output transistors mounted on a big heatsink...
...
Next I increase the current to 1 amp. After which something must have blown. I didn't hear or smell anything but I could't read any voltage on the output terminals any longer.

The schematic shows current sources as a gain control element, so the stage gain will be very high. As you laid this out on a breadboard, there are extra capacitances not shown on the schematic taking effect.

I'm thinking that one of your transistors started oscillating at an ultrasonic frequency (or higher) and burned itself out, or forced downstream transistors to burn themselves out.

I had the same problem when modifying one of those old SWTP amps to improve its TIM distortion (because it had high stage gain and high feedback to lower THD).

For a couple minutes, I thought I had vastly improved the performance as I was hearing nuances and details I had never heard before. Turns out, the outputs were just ringing with every impulse.

Then it stopped putting any sound out altogether and I noticed that the power transistors were very hot.

I would suggest replacing the transistors and running your tests with scope leads on the positive power supply and the low side of R10 (AC coupling for both, with 1V/div and 10-100 uS/div settings).

If you see any indication of non-audio signals present, then beef up your supply capacitance (which may possibly help) or change the layout of your board (I'd put it on a PCB with ground planes).


Jeff
 
Ok, your feedback is great!!! Thank you all for jumping in and sharing ideas.

After reading several of the comments above I went back and double checked everything I was doing and I wasn't doing everything right.

Below is a picture of how I have things grounded. I know you probably can't tell much other than I'm not very good at soldering. ;) I have the ground lead coming from my power supply clipped to the big bare copper ground wire running the length of the board.

When I attach the negative probe of my DMM to the ground post on the power supply and check voltages I can measure -22v on the negative rail but only 0.4v on the positive rail.

When I disconnect the ground clip and measure across VE+ and VE- I get 22v.

An externally hosted image should be here but it was not working when we last tested it.


Obviously I've done something wrong...
 
Sorry to say this but I see a disaster waiting to happen or rather happened. In my very humble and honest opinion, I would urge you to find a PCB on eBay for this amp and build it properly.

Your heat sink is good for this amp. The application is scary :eek:

@Jeff, I doubt stay capacitance is to be a problem here

Edit: By the sound of it your lab power supply may gone into current limiting (maybe a good thing) if there was a fault and could have saved the components.
 
Last edited:
When I attach the negative probe of my DMM to the ground post on the power supply and check voltages I can measure -22v on the negative rail but only 0.4v on the positive rail.

When I disconnect the ground clip and measure across VE+ and VE- I get 22v.

Based on what you've measured, I'd say you burned out Q2.

You've got way too much space between your components for a circuit with high stage gain that relies on negative feedback to establish operating points.

Find (or make) a PCB for this amplifier.


Jeff
 
Administrator
Joined 2007
Paid Member
When I attach the negative probe of my DMM to the ground post on the power supply and check voltages I can measure -22v on the negative rail but only 0.4v on the positive rail.

When I disconnect the ground clip and measure across VE+ and VE- I get 22v.


Obviously I've done something wrong...

Unless I've missed it somewhere, what is the power supply you are using ?
 
HAHAHA You guys crack me up!!!! :)

I agree... time to find a nice PC board to build this amp on.

Nico thanks for the link to the PC board. Does anyone else have any recommendations for any tried and true JLH96 two output transistor board?

Mooly here is the power supply I've been using:

An externally hosted image should be here but it was not working when we last tested it.


Many thanks again to all of you who took time to comment. This is a great group and your help was very much appreciated.

While I'm waiting for my new PC boards to arrive I guess I'll get started on the capacitance multiplier power supply I intend to use. Bet you guys can't wait to see that!! :)

Bruce